THE theory of the potential distribution in a long cable when the head end is insulated after being maintained for some time at a steady potential has been given recently.1 It was pointed out that the reasoning would apply equally well to the case of a long uniform bar which is allowed to cool after one end has been heated. The experimental confirmation of the theory is not easy in the electrical case owing to the fact that it is not possible to tap a long laid cable at intervals for the observation of the potential, and artificial lines are not available with the leakance sufficiently uniformly distributed. Moreover, the changes are very rapid, and in the practical example which was worked out in the above-mentioned paper, all transients had practically died away after the expiry of two seconds. In the heat analogue, on the other hand, the changes are very much slower and the necessary apparatus (an ordinary ‘Forbes bar’) available in almost any laboratory.